Over 40 percent of patients living with diabetes are over 65 years of age, yet very little is known about the risks and benefits of diabetes care in an aging population. This vulnerable population is more heterogeneous than younger patients in terms of functional status, comorbidities and life expectancy, and may need more individualized management of glucose, blood pressure, and cholesterol levels. The special concern that motivates this study (especially since excess deaths led to the discontinuation of the intensive glucose-lowering arm in the ACCORD trial) is the possibility that the risks or burden of intensive and comprehensive diabetes care may outweigh the benefits for some older patients. Geriatric diabetes care guidelines currently recommend that providers consider 5-year life expectancy, risks of hypoglycemia, and geriatric syndromes (e.g., falls, cognitive decline) when setting diabetes treatment goals. While these new recommendations have a clinical rationale, they have not been formally evaluated with data from contemporary epidemiological studies, they lack validated prognostic tools for risk-stratification (predictive 5-year survival models), and their policy implications for quality assessment and health disparities have not been explored. Moreover, little is known about ethnic differences in the natural history of diabetes and their effect on care and outcomes in the elderly. We propose a prospective (2005-2014) examination of a large contemporary, multi-ethnic cohort of ~112,000 older (e60 years of age) diabetes patients identified from the Kaiser Permanente Northern California Diabetes Registry, and sub-studies in a cohort of survey responders (N=9,763). We will: i) characterize the current state of diabetes care management by health status, evaluate quality of life, and estimate rates of traditional complications, hypoglycemia, geriatric syndromes and mortality; ii) describe the interrelationships of hypoglycemia and geriatric syndromes; iii) evaluate the effects of antihyperglycemic therapies and polypharmacy on hypoglycemia, geriatric syndromes, and mortality; iv) create a generalized prediction model for 5-year, all-cause mortality and evaluate the performance of existing prognostic mortality prediction models; and v) explore the health policy implications of the widespread adoption of geriatric diabetes guidelines. This multi-institution, prospective study will expand our understanding of the dynamics of health care and outcomes among elderly diabetes patients. Study findings will provide important insight into the validity and implications of geriatric guidelines, and help ensure that quality improvement efforts for geriatric diabetes care are not at odds with efforts to improve diabetes outcomes and will not increase health disparities within health systems.